How higher goals are constructed and collapse under stress: A hierarchical Bayesian control systems perspective

Area of Science:

• Computational neuroscience
• Systems biology
• Theoretical psychology

Background:

• Organisms can be modeled as complex control systems.
• Hierarchical structures are prevalent in biological systems.
• Bayesian inference plays a role in biological decision-making.

Purpose of the Study:

• To propose a model of organisms as hierarchical Bayesian control systems.
• To investigate the role of network structure (small-world, bow-tie) in these systems.
• To explore the implications of goal hierarchies for organismal behavior and health.

Main Methods:

• Developing a theoretical framework for hierarchical Bayesian control.
• Analyzing network properties of biological control systems.
• Modeling goal hierarchy decomposition under stress.
• Connecting model dynamics to organismal outcomes like disease and personality.

Main Results:

• Organisms are modeled as hierarchical Bayesian control systems with specific network structures.
• These systems generate deep goal hierarchies, including self, social, and normative models.
• Stress causes top-down decomposition of goal hierarchies, prioritizing short-term, self-referential goals.
• Goal hierarchy collapse correlates with increased system entropy, serving as an early warning for critical tipping points.
• In humans, goal hierarchy maturation relates to personality development; failure indicates deficits.

Conclusions:

• Hierarchical Bayesian control systems provide a unifying framework for understanding organismal regulation.
• The proposed model explains stress-induced behavioral shifts and links to psychopathology.
• Increased system entropy due to goal hierarchy collapse is a potential biomarker for disease or mortality.
• The framework offers testable hypotheses for empirical validation in biological and psychological research.